[cma]

So are you saying if the government wanted to subsidize chip development and had these choices:

A) Make resistive element chips out of wafers and all the normal complex chip patterning, using up the fab side of things and wafer space

B) Subsidize custom chip molecular simulation accelerators for cancer research, by providing the fab side of things and wafer space to design researchers or companies

A is as good as B?

[Dylan16807]

No. Why do you think I'm saying that?

I'm saying that over the course of several years, A is better for chip advancement and availability than nothing at all. And in the real world it's A+B vs. just B, the same amount of B in both scenarios.

I'm not aware of Bitcoin reducing any cancer research budgets anywhere. Are you?

[cma]

This would be like the government proposing buying up a big portion of argricultural tractors and running hours on them until their engines wear out, in order to help the economy make more tractors. Cash for clunkers without the environmental rationale.

[Dylan16807]

That would be bad because of the CO2 emissions and because it's a huge waste of government money.

It would not be bad for long-term tractor availability, if they kept up big purchases for many years. Especially if tractors were a hotbed of research and improving by orders of magnitude.

Or to put it another way: It's a hugely inefficient way of boosting tractors, but it will in fact boost tractors. It's not a good idea overall, but it will have some positives.